Frozen-section assessment of sentinel lymph nodes in breast cancer: diagnostic concordance and axillary management in a South Asian cohort

Intraoperative frozen-section assessment of sentinel lymph nodes (SLNs) may facilitate immediate axillary treatment; however, its value is diminishing as the indications for both SLN biopsy and axillary lymph-node dissection (ALND) narrow. The clinically relevant question is whether the result provides actionable information rather than diagnostic accuracy alone. This study evaluated patient-level concordance and the association between frozen-section findings and axillary management in a tertiary-care breast cancer cohort in Pakistan. This retrospective cohort included 322 patients who underwent SLN biopsy with intraoperative frozen-section assessment at a tertiary-care academic center in Islamabad, Pakistan. Frozen-section findings were compared with the recorded final axillary nodal status, which could include non-sentinel nodes after ALND; estimates were therefore interpreted as patient-level concordance rather than same-node permanent histopathological validation. Associations with subsequent axillary management were examined, and logistic regression identified factors associated with recorded final nodal positivity. Among 309 patients with paired data, frozen-section assessment showed 74.3% sensitivity, 99.5% specificity, 98.7% positive predictive value, 88.8% negative predictive value, and 91.3% overall concordance against the recorded final nodal status. Twenty-six node-positive patients had negative frozen sections. ALND was performed in 59 of 76 frozen-section-positive patients and 22 of 238 frozen-section-negative patients. Because procedural timing and case-specific indications were not consistently recorded, the number in whom frozen section directly changed intraoperative management could not be determined. T2-T4 disease independently predicted recorded final nodal positivity. Frozen-section assessment yielded highly reliable positive findings but moderate sensitivity against the recorded final nodal status. Frozen section positivity was associated with ALND; however, this association did not establish a direct intraoperative management change. Selective use is most defensible when a positive result has a predefined same-session therapeutic consequence, whereas definitive pathological assessment remains essential after a negative result is obtained.